Smoother class: The smoother class is designed for to be used in conjunction with multigrid. The action of the smoother should reduce the high frequency errors. These methods are inefficient as stand-alone solvers. More...
#include <iterative_linear_solver.h>
Inheritance diagram for oomph::Smoother:
Public Member Functions | |
| Smoother () | |
| Empty constructor. | |
| virtual | ~Smoother () |
| Virtual empty destructor. | |
| virtual void | smoother_solve (const DoubleVector &rhs, DoubleVector &result)=0 |
| The smoother_solve function performs fixed number of iterations on the system A*result=rhs. The number of (smoothing) iterations is the same as the max. number of iterations in the underlying IterativeLinearSolver class. Note that the result vector MUST NOT re-initialised to zero (as it would typically be when the Smoother is called as a iterative linear solver). | |
| virtual void | smoother_setup (DoubleMatrixBase *matrix_pt)=0 |
| Set up the smoother for the matrix specified by the pointer. | |
| template<typename MATRIX > | |
| void | check_validity_of_solve_helper_inputs (MATRIX *const &matrix_pt, const DoubleVector &rhs, DoubleVector &solution, const double &n_dof) |
| Self-test to check that all the dimensions of the inputs to solve helper are consistent and everything that needs to be built, is. | |
| Public Member Functions inherited from oomph::IterativeLinearSolver | |
| IterativeLinearSolver () | |
| Constructor: Set (default) trivial preconditioner and set defaults for tolerance and max. number of iterations. | |
| IterativeLinearSolver (const IterativeLinearSolver &)=delete | |
| Broken copy constructor. | |
| void | operator= (const IterativeLinearSolver &)=delete |
| Broken assignment operator. | |
| virtual | ~IterativeLinearSolver () |
| Destructor (empty) | |
| Preconditioner *& | preconditioner_pt () |
| Access function to preconditioner. | |
| Preconditioner *const & | preconditioner_pt () const |
| Access function to preconditioner (const version) | |
| double & | tolerance () |
| Access to convergence tolerance. | |
| unsigned & | max_iter () |
| Access to max. number of iterations. | |
| virtual unsigned | iterations () const =0 |
| Number of iterations taken. | |
| void | enable_doc_convergence_history () |
| Enable documentation of the convergence history. | |
| void | disable_doc_convergence_history () |
| Disable documentation of the convergence history. | |
| void | open_convergence_history_file_stream (const std::string &file_name, const std::string &zone_title="") |
| Write convergence history into file with specified filename (automatically switches on doc). Optional second argument is a string that can be used (as a zone title) to identify what case we're running (e.g. what combination of linear solver and preconditioner or parameter values are used). | |
| void | close_convergence_history_file_stream () |
| Close convergence history output stream. | |
| double | jacobian_setup_time () const |
| returns the time taken to assemble the jacobian matrix and residual vector | |
| double | linear_solver_solution_time () const |
| return the time taken to solve the linear system | |
| virtual double | preconditioner_setup_time () const |
| returns the the time taken to setup the preconditioner | |
| void | enable_setup_preconditioner_before_solve () |
| Setup the preconditioner before the solve. | |
| void | disable_setup_preconditioner_before_solve () |
| Don't set up the preconditioner before the solve. | |
| void | enable_error_after_max_iter () |
| Throw an error if we don't converge within max_iter. | |
| void | disable_error_after_max_iter () |
| Don't throw an error if we don't converge within max_iter (default). | |
| void | enable_iterative_solver_as_preconditioner () |
| Enables the iterative solver be used as preconditioner (when calling the solve method it bypass the setup solver method – currently only used by Trilinos solver —) | |
| void | disable_iterative_solver_as_preconditioner () |
| Disables the iterative solver be used as preconditioner (when calling the solve method it bypass the setup solver method – currently only used by Trilinos solver —) | |
| Public Member Functions inherited from oomph::LinearSolver | |
| LinearSolver () | |
| Empty constructor, initialise the member data. | |
| LinearSolver (const LinearSolver &dummy)=delete | |
| Broken copy constructor. | |
| void | operator= (const LinearSolver &)=delete |
| Broken assignment operator. | |
| virtual | ~LinearSolver () |
| Empty virtual destructor. | |
| void | enable_doc_time () |
| Enable documentation of solve times. | |
| void | disable_doc_time () |
| Disable documentation of solve times. | |
| bool | is_doc_time_enabled () const |
| Is documentation of solve times enabled? | |
| bool | is_resolve_enabled () const |
| Boolean flag indicating if resolves are enabled. | |
| virtual void | enable_resolve () |
| Enable resolve (i.e. store matrix and/or LU decomposition, say) Virtual so it can be overloaded to perform additional tasks. | |
| virtual void | disable_resolve () |
| Disable resolve (i.e. store matrix and/or LU decomposition, say) This function simply resets an internal flag. It's virtual so it can be overloaded to perform additional tasks such as cleaning up memory that is only required for the resolve. | |
| virtual void | solve (Problem *const &problem_pt, DoubleVector &result)=0 |
| Solver: Takes pointer to problem and returns the results vector which contains the solution of the linear system defined by the problem's fully assembled Jacobian and residual vector. | |
| virtual void | solve (DoubleMatrixBase *const &matrix_pt, const DoubleVector &rhs, DoubleVector &result) |
| Linear-algebra-type solver: Takes pointer to a matrix and rhs vector and returns the solution of the linear system. | |
| virtual void | solve (DoubleMatrixBase *const &matrix_pt, const Vector< double > &rhs, Vector< double > &result) |
| Linear-algebra-type solver: Takes pointer to a matrix and rhs vector and returns the solution of the linear system. | |
| virtual void | solve_transpose (Problem *const &problem_pt, DoubleVector &result) |
| Solver: Takes pointer to problem and returns the results vector which contains the solution of the linear system defined by the problem's fully assembled Jacobian and residual vector (broken virtual). | |
| virtual void | solve_transpose (DoubleMatrixBase *const &matrix_pt, const DoubleVector &rhs, DoubleVector &result) |
| Linear-algebra-type solver: Takes pointer to a matrix and rhs vector and returns the solution of the linear system. | |
| virtual void | solve_transpose (DoubleMatrixBase *const &matrix_pt, const Vector< double > &rhs, Vector< double > &result) |
| Linear-algebra-type solver: Takes pointer to a matrix and rhs vector and returns the solution of the linear system. | |
| virtual void | resolve (const DoubleVector &rhs, DoubleVector &result) |
| Resolve the system defined by the last assembled jacobian and the rhs vector. Solution is returned in the vector result. (broken virtual) | |
| virtual void | resolve_transpose (const DoubleVector &rhs, DoubleVector &result) |
| Solver: Resolve the system defined by the last assembled jacobian and the rhs vector. Solution is returned in the vector result. (broken virtual) | |
| virtual void | clean_up_memory () |
| Empty virtual function that can be overloaded in specific linear solvers to clean up any memory that may have been allocated (e.g. when preparing for a re-solve). | |
| virtual void | enable_computation_of_gradient () |
| function to enable the computation of the gradient required for the globally convergent Newton method | |
| void | disable_computation_of_gradient () |
| function to disable the computation of the gradient required for the globally convergent Newton method | |
| void | reset_gradient () |
| function to reset the size of the gradient before each Newton solve | |
| void | get_gradient (DoubleVector &gradient) |
| function to access the gradient, provided it has been computed | |
| Public Member Functions inherited from oomph::DistributableLinearAlgebraObject | |
| DistributableLinearAlgebraObject () | |
| Default constructor - create a distribution. | |
| DistributableLinearAlgebraObject (const DistributableLinearAlgebraObject &matrix)=delete | |
| Broken copy constructor. | |
| void | operator= (const DistributableLinearAlgebraObject &)=delete |
| Broken assignment operator. | |
| virtual | ~DistributableLinearAlgebraObject () |
| Destructor. | |
| LinearAlgebraDistribution * | distribution_pt () const |
| access to the LinearAlgebraDistribution | |
| unsigned | nrow () const |
| access function to the number of global rows. | |
| unsigned | nrow_local () const |
| access function for the num of local rows on this processor. | |
| unsigned | nrow_local (const unsigned &p) const |
| access function for the num of local rows on this processor. | |
| unsigned | first_row () const |
| access function for the first row on this processor | |
| unsigned | first_row (const unsigned &p) const |
| access function for the first row on this processor | |
| bool | distributed () const |
| distribution is serial or distributed | |
| bool | distribution_built () const |
| if the communicator_pt is null then the distribution is not setup then false is returned, otherwise return true | |
| void | build_distribution (const LinearAlgebraDistribution *const dist_pt) |
| setup the distribution of this distributable linear algebra object | |
| void | build_distribution (const LinearAlgebraDistribution &dist) |
| setup the distribution of this distributable linear algebra object | |
Protected Attributes | |
| bool | Use_as_smoother |
| When a derived class object is being used as a smoother in the MG solver (or elsewhere) the residual norm does not need to be calculated because we're simply performing a fixed number of (smoothing) iterations. This boolean is used as a flag to indicate that the IterativeLinearSolver (which this class is by inheritance) is supposed to act in this way. | |
| Protected Attributes inherited from oomph::IterativeLinearSolver | |
| bool | Doc_convergence_history |
| Flag indicating if the convergence history is to be documented. | |
| std::ofstream | Output_file_stream |
| Output file stream for convergence history. | |
| double | Tolerance |
| Convergence tolerance. | |
| unsigned | Max_iter |
| Maximum number of iterations. | |
| Preconditioner * | Preconditioner_pt |
| Pointer to the preconditioner. | |
| double | Jacobian_setup_time |
| Jacobian setup time. | |
| double | Solution_time |
| linear solver solution time | |
| double | Preconditioner_setup_time |
| Preconditioner setup time. | |
| bool | Setup_preconditioner_before_solve |
| indicates whether the preconditioner should be setup before solve. Default = true; | |
| bool | Throw_error_after_max_iter |
| Should we throw an error instead of just returning when we hit the max iterations? | |
| bool | Use_iterative_solver_as_preconditioner |
| Use the iterative solver as preconditioner. | |
| bool | First_time_solve_when_used_as_preconditioner |
| When the iterative solver is used a preconditioner then we call the setup of solver method only once (the first time the solve method is called) | |
| Protected Attributes inherited from oomph::LinearSolver | |
| bool | Enable_resolve |
| Boolean that indicates whether the matrix (or its factors, in the case of direct solver) should be stored so that the resolve function can be used. | |
| bool | Doc_time |
| Boolean flag that indicates whether the time taken. | |
| bool | Compute_gradient |
| flag that indicates whether the gradient required for the globally convergent Newton method should be computed or not | |
| bool | Gradient_has_been_computed |
| flag that indicates whether the gradient was computed or not | |
| DoubleVector | Gradient_for_glob_conv_newton_solve |
| DoubleVector storing the gradient for the globally convergent Newton method. | |
Additional Inherited Members | |
| Protected Member Functions inherited from oomph::DistributableLinearAlgebraObject | |
| void | clear_distribution () |
| clear the distribution of this distributable linear algebra object | |
| Static Protected Attributes inherited from oomph::IterativeLinearSolver | |
| static IdentityPreconditioner | Default_preconditioner |
| Default preconditioner: The base class for preconditioners is a fully functional (if trivial!) preconditioner. | |
Detailed Description
Smoother class: The smoother class is designed for to be used in conjunction with multigrid. The action of the smoother should reduce the high frequency errors. These methods are inefficient as stand-alone solvers.
Definition at line 549 of file iterative_linear_solver.h.
Constructor & Destructor Documentation
◆ Smoother()
|
inline |
Empty constructor.
Definition at line 553 of file iterative_linear_solver.h.
◆ ~Smoother()
|
inlinevirtual |
Virtual empty destructor.
Definition at line 556 of file iterative_linear_solver.h.
Member Function Documentation
◆ check_validity_of_solve_helper_inputs()
| void oomph::Smoother::check_validity_of_solve_helper_inputs | ( | MATRIX *const & | matrix_pt, |
| const DoubleVector & | rhs, | ||
| DoubleVector & | solution, | ||
| const double & | n_dof | ||
| ) |
Self-test to check that all the dimensions of the inputs to solve helper are consistent and everything that needs to be built, is.
Self-test to be called inside solve_helper to ensure that all inputs are consistent and everything that needs to be built, is.
Definition at line 1029 of file iterative_linear_solver.cc.
References oomph::DistributableLinearAlgebraObject::distributed().
◆ smoother_setup()
|
pure virtual |
Set up the smoother for the matrix specified by the pointer.
Implemented in oomph::GS< MATRIX >, oomph::GS< CRDoubleMatrix >, and oomph::DampedJacobi< MATRIX >.
◆ smoother_solve()
|
pure virtual |
The smoother_solve function performs fixed number of iterations on the system A*result=rhs. The number of (smoothing) iterations is the same as the max. number of iterations in the underlying IterativeLinearSolver class. Note that the result vector MUST NOT re-initialised to zero (as it would typically be when the Smoother is called as a iterative linear solver).
Implemented in oomph::GS< MATRIX >, oomph::GS< CRDoubleMatrix >, and oomph::DampedJacobi< MATRIX >.
Member Data Documentation
◆ Use_as_smoother
|
protected |
When a derived class object is being used as a smoother in the MG solver (or elsewhere) the residual norm does not need to be calculated because we're simply performing a fixed number of (smoothing) iterations. This boolean is used as a flag to indicate that the IterativeLinearSolver (which this class is by inheritance) is supposed to act in this way.
Definition at line 585 of file iterative_linear_solver.h.
Referenced by oomph::GS< MATRIX >::smoother_solve(), oomph::GS< CRDoubleMatrix >::smoother_solve(), oomph::DampedJacobi< MATRIX >::smoother_solve(), oomph::GS< MATRIX >::solve(), oomph::GS< CRDoubleMatrix >::solve(), and oomph::DampedJacobi< MATRIX >::solve().
The documentation for this class was generated from the following files: